Glioma biopsies Classification Using Raman Spectroscopy and Machine Learning Models on Fresh Tissue Samples

Simple Summary Raman Spectroscopy (RS) is an optical technique able to determine biochemical differences within a biological tissue, yet there are few reports of its application in fresh tissue. This potential could be relevant to detect glioma infiltration and to guide surgery to achieve a total resection. We deployed RS with machine learning algorithms in a surgical scenario to probe the diagnostic accuracy ex-vivo in discriminating between normal-appearing and fresh neoplastic tissue in patients undergoing glioma resection. Analyzing 3450 spectra from 63 samples, we identified 19 novel RS shifts. These 19 novel shifts of known biological relevance were included in the analytical workflow, leading to 83% and 82% accuracy and precision, respectively, in discriminating between normal-appearing and neoplastic tissue. This study further supported the translational development of real-time tissue analysis with RS in oncological brain surgery and yielded novel findings in the biochemical features of the brain tissue affected by glioma. Identifying tumor cells infiltrating normal-appearing brain tissue is critical to achieve a total glioma resection. Raman spectroscopy (RS) is an optical technique with potential for real-time glioma detection. Most RS reports are based on formalin-fixed or frozen samples, with only a few studies deployed on fresh untreated tissue. We aimed to probe RS on untreated brain biopsies exploring novel Raman bands useful in distinguishing glioma and normal brain tissue. Sixty-three fresh tissue biopsies were analyzed within few minutes after resection. A total of 3450 spectra were collected, with 1377 labelled as Healthy and 2073 as Tumor. Machine learning methods were used to classify spectra compared to the histo-pathological standard. The algorithms extracted information from 60 different Raman peaks identified as the most representative among 135 peaks screened. We were able to distinguish between tumor and healthy brain tissue with accuracy and precision of 83% and 82%, respectively. We identified 19 new Raman shifts with known biological significance. Raman spectroscopy was effective and accurate in discriminating glioma tissue from healthy brain ex-vivo in fresh samples. This study added new spectroscopic data that can contribute to further develop Raman Spectroscopy as an intraoperative tool for in-vivo glioma detection.

Automated summary

Raman spectroscopy has the potential for real-time glioma detection, especially in fresh tissue. This study utilized machine learning algorithms to accurately distinguish between normal and tumor tissue in surgical scenarios, contributing new spectral data for intraoperative glioma detection in neurosurgery.